Download ogle-lmc-ecl-11893: the discovery of a long-period

The Astrophysical Journal, 788:41 (4pp), 2014 June 10
C 2014.
doi:10.1088/0004-637X/788/1/41
The American Astronomical Society. All rights reserved. Printed in the U.S.A.
OGLE-LMC-ECL-11893: THE DISCOVERY OF A LONG-PERIOD ECLIPSING
BINARY WITH A CIRCUMSTELLAR DISK
1
Subo Dong1 , Boaz Katz2,11 , Jose L. Prieto3,12 , Andrzej Udalski4,13 , Szymon Kozlowski4,13 ,
R. A. Street5,14 , D. M. Bramich6,14 , Y. Tsapras5,7,14 , M. Hundertmark8,14 , C. Snodgrass9,14 ,
K. Horne8,14 , M. Dominik8,14,15 , and R. Figuera Jaimes8,10,14
Kavli Institute for Astronomy and Astrophysics, Peking University, Yi He Yuan Road 5, Hai Dian District, Beijing 100871, China
2 Institute for Advanced Study, 1 Einstein Drive, Princeton, NJ 08544, USA
3 Department of Astrophysical Sciences, Princeton University, 4 Ivy Lane, Peyton Hall, Princeton, NJ 08544, USA
4 Warsaw University Observatory, Al. Ujazdowskie 4, 00-478 Warszawa, Poland
5 Las Cumbres Observatory Global Telescope Network, 6740 Cortona Drive, suite 102, Goleta, CA 93117, USA
6 Qatar Environment and Energy Research Institute, Qatar Foundation, Tornado Tower, Floor 19, P.O. Box 5825, Doha, Qatar
7 School of Physics and Astronomy, Queen Mary University of London, Mile End Road, London E1 4NS, UK
8 SUPA, School of Physics and Astronomy, University of St Andrews, North Haugh, St Andrews KY16 9SS, UK
9 Max Planck Institute for Solar System Research, Justus-von-Liebig-Weg 3, D-37077 Göttingen, Germany
10 European Southern Observatory, Karl-Schwarzschild-Straße 2, D-85748 Garching bei München, Germany
Received 2014 March 5; accepted 2014 April 15; published 2014 May 20
ABSTRACT
We report the serendipitous discovery of a disk–eclipse system OGLE-LMC-ECL-11893. The eclipse occurs with
a period of 468 days, a duration of about 15 days, and a deep (up to ΔmI ≈ 1.5), peculiar, and asymmetric profile.
A possible origin of such an eclipse profile involves a circumstellar disk. The presence of the disk is confirmed
by the H-α line profile from the follow-up spectroscopic observations, and the star is identified as Be/Ae type.
Unlike the previously known disk–eclipse candidates, the eclipses of OGLE-LMC-ECL-11893 retain the same
shape throughout the span of ∼17 yr (13 orbital periods), indicating no measurable orbital precession of the disk.
Key words: binaries: eclipsing – circumstellar matter – stars: individual (OGLE-LMC-ECL-11893) – stars:
variables: T Tauri – Herbig Ae/Be
Online-only material: color figures, machine-readable table
the duration of EE Cephei is also similar (∼30 days). The
ingress and egress of OGLE-LMC-ECL-11893 are steeper
than those of EE Cephei.
2. The eclipse is deep, with a maximal depth of ∼1.5 mag or
equivalent, and 75% of the I-band light is occulted. This
is comparable to that of EE Cephei, whose eclipse depths
vary between ∼0.5 and 2.0 mag (Gałan et al. 2012).
3. The shape of the eclipse does not change with time. The
nine eclipses spanning 17 yr detected so far have exactly
the same profile within measurement uncertainties. This
is in contrast with EE Cephei and three other well-known
disk–eclipse events (ε Aurigae, OGLE-LMC-ECL-17782,
and KH 15D), whose eclipse depths vary from one orbital
period to another. These variations are generally attributed
to the precessions of the disks.
1. IDENTIFICATION OF THE PECULIAR
ECLIPSES OF OGLE-LMC-ECL-11893
The OGLE LMC eclipsing binary (EB) catalog (Graczyk
et al. 2011) reported the discovery of 26121 EBs toward
the LMC. In 2011 November, while searching the catalog
for long-period (period P > 100 days), highly eccentric
binaries (Dong et al. 2013), we serendipitously identified a
very unusual EB OGLE-LMC-ECL-11893 (R.A.: 05:17:21.17
decl.: −69:00:55.7; J2000) by visual inspection. The periodfolded I-band light curve shows an eclipse with a peculiar shape
(Figure 1). The data are taken from phases II and III of the OGLE
survey (Udalski et al. 1997, 2008) and span 12 yr (further data
from OGLE-IV survey are discussed in Section 2). The period is
468.124 days, and the baseline I-band magnitude is about 17.6.
Each eclipse occurrence is shown with a different color. The
eclipse bears some close similarity to EE Cephei, an eclipsing
Be star with a circumstellar disk (Gałan et al. 2012), but also
has some unique characteristics. In the following, we list several
striking features of the OGLE-LMC-ECL-11893 eclipses.
The similarities between OGLE-LMC-ECL-11893 and EE
Cephei hint that they have similar physical origins, possibly
disk-eclipsing events, as suggested by Gałan et al. (2012) for
EE Cephei.
1. The eclipse profile, which spans about 15 days, has an
unusual asymmetric shape that cannot be produced by a
spherical occulting star. The shape of the profile bears a
close similarity to EE Cephei during its 2008/9 eclipse
(e.g., see the right panel of Figure 3 in Gałan et al. 2012) and
2. FOLLOW-UP OBSERVATIONS
We obtained two optical spectra of OGLE-LMC-ECL-11893,
first on 2011 December 23 with IMACS f/2 (Dressler et al.
2011) mounted on the Magellan Baade 6.5 m and the second
on 2013 January 7 with the Magellan Echellette spectrograph
(MagE; Marshall et al. 2008). For the IMACS f/2 spectrum,
we used the 300 l mm−1 grism and 0. 7 slit, which gives
resolution R = 1800 and wavelength coverage 3700–9500 Å.
We show the IMACS spectrum of the star in the upper panel of
Figure 2. MagE with 1 slit gives R = 4800 and wavelength
11
John N. Bahcall Fellow.
Carnegie-Princeton Fellow.
13 The OGLE Collaboration.
14 The RoboNet Collaboration.
15 Royal Society University Research Fellow.
12
1
The Astrophysical Journal, 788:41 (4pp), 2014 June 10
Dong et al.
Figure 1. Period-folded OGLE I-band light curve of OGLE-LMC-ECL-11893
with respect to HJD0 = 2,454,245.5663. Photometry from different orbital
periods is shown in different colors. The observations by OGLE-II and OGLE-III
span ∼12 yr and contain nine eclipse occurrences. The peculiar and asymmetric
eclipse profile is unchanged within measurement uncertainties during the span
of observations.
(A color version of this figure is available in the online journal.)
coverage 3300–10,000 Å. The IMACS spectrum was reduced
with standard techniques in IRAF. The MagE spectrum was
reduced with the Carnegie pipeline from D. Kelson.
Using the Ulyss package, which constrains stellar parameters by fitting stellar spectra with the Elodie spectroscopy
library (Koleva et al. 2009), we find that the spectrum is
consistent with that of a late B/early A-type giant with
Teff ∼ 10,000 K. The RV of the star measured from the
MagE spectrum is 270 km s−1 , consistent with the radial velocity (RV) of the LMC. In both spectra, we find
H-α line profiles characteristic of circumstellar disks (Silaj et al.
2010), confirming the existence of the disk indicated by the
eclipse and that the system is of Be/Ae type. The H-α profile
in the MagE spectrum is shown in the lower panel of Figure 2.
The H-α profile is similar to that of ε Aurigae (Silaj et al. 2010),
typical for a Be star observed edge-on (see Figure 3 in Slettebak 1979). ε Aurigae is a disk-eclipsing event for which the disk
was directly confirmed by infrared interferometry (Kloppenborg
et al. 2010).
Multi-band photometry obtained using the 1 m telescopes
(Domes A and B) at LCOGT’s site in Chile (CTIO) during 2013
February and March (at the early commissioning stages of the
telescopes) are shown in Figure 3 and given in Table 1. The
observations were conducted through the Observation Control
interface developed for the RoboNet project, taken in semirobotic mode as the underlying software was in the process of
being upgraded to integrate with the new 1 m network at the time.
The image data were preprocessed using LCOGT’s Standard
Pipeline, which is based around ORAC-DR, after which the
RoboNet pipeline (based on DanDIA by Bramich 2008) was
used for difference image analysis photometry.
The different bands (filled and empty dots) are compared to
the folded I and V bands OGLE II, III, and IV photometry (empty
Figure 2. Top: spectrum of OGLE-LMC-ECL-11893 taken by IMACS (black).
The best-fit model from Ulyss (red) suggests it is a late B/early A-type giant.
Bottom: the H-α line profile in the MagE spectrum that is characteristic of
Ae/Be stars confirms the existence of a circumstellar disk (Silaj et al. 2010).
The same profile also clearly shows in the IMACS spectrum, though the H-α
region is near the edge of the CCD. The H-α profile is typical for an edge-on
Be star, similar to that of ε Aurigae (see bottom left panel of Figure 8 on page
241 in Silaj et al. 2010), for which the disk was directly confirmed by infrared
interferometer imaging (Kloppenborg et al. 2010).
(A color version of this figure is available in the online journal.)
squares and filled and empty pentagons). Given the photometric
uncertainties in the LCOGT data, there is reasonable agreement
between the LCOGT data and the OGLE data for the V and
I bands where they could be directly compared. While most
bands have a roughly similar profile, the eclipse depth seems
to increase for bluer bands (from I, R, V, to B). In particular,
the B band seems to have a consistently deeper eclipse and the
two B-band measurements during HJD 2,456,353, in the deepest
point of the eclipse, seem to be particularly deep compared to
2
The Astrophysical Journal, 788:41 (4pp), 2014 June 10
Dong et al.
Table 1
Time-series B, V, R, and I Photometry for OGLE-LMC-ECL-11893 from the LCOGT Network of Telescopes
Telescope
ID
lsc-doma-1m0-05-kb78
lsc-doma-1m0-05-kb78
..
.
lsc-doma-1m0-05-kb78
lsc-doma-1m0-05-kb78
.
..
lsc-domb-1m0-09-kb73
lsc-domb-1m0-09-kb73
Filter
HJD
(days)
mins
(mag)
σm
(mag)
fref
(ADU s−1 )
σref
(ADU s−1 )
fdiff
(ADU s−1 )
σdiff
(ADU s−1 )
p
B
B
..
.
V
V
.
..
I
I
2,456,343.58641
2,456,343.59010
..
.
2,456,343.60545
2,456,343.68000
.
..
2,456,364.58433
2,456,364.58801
21.502
21.404
..
.
21.035
21.011
.
..
20.885
20.923
0.034
0.032
..
.
0.021
0.026
.
..
0.019
0.020
69.323
69.323
..
.
38.004
38.004
.
..
43.815
43.815
1.879
1.879
..
.
3.224
3.224
.
..
2.107
2.107
−43.868
−41.498
..
.
0.599
1.451
.
..
0.426
−1.077
0.773
0.804
..
.
0.819
0.976
.
..
0.787
0.786
0.9913
0.9908
..
.
1.0731
1.0451
.
..
1.0018
1.0024
Notes. The instrumental mins magnitudes are listed in Column 4 corresponding to the LCOGT site/dome/telescope/instrument code, filter,
and epoch of mid-exposure listed in Columns 1–3, respectively. The uncertainty on mins is listed in Column 5. For completeness, we also
list the quantities fref , fdiff , and p in Columns 6, 8, and 10, along with the uncertainties σref and σdiff in Columns 7 and 9. Note that
mins (t) = 25–2.5 log(fref + (fdiff (t)/p(t))), where t is time. Note that these magnitudes are not calibrated to standard systems and they
should be used only for studying the relative flux changes. The calibrated magnitudes of the system at the baseline are: I = 17.55 ± 0.03,
V = 17.74 ± 0.02, and B = 17.95 ± 0.04.
(This table is available in its entirety in a machine-readable form in the online journal. A portion is shown here for guidance regarding its form
and content.)
able as it will be too close to the Sun. We encourage intense
multi-band photometric and spectroscopic monitoring during
the eclipse. In addition, long-term RV monitoring will be useful
to determine the orbital parameters and the mass of the eclipsing
component.
3. DISCUSSION
Assuming a circular orbit, the size of an eclipsing region
is of the order of: D ∼ 2π aΔt/P = Δt(2π GMtot /P )1/3 ∼
0.35(Mtot /3M )1/3 AU for eclipse duration Δt ∼ 15 days and
P = 468 days, where Mtot is the total mass of the system and a
is the semi-major axis.
Three well-known periodic disk–eclipse EB candidates are
ε Aurigae (P = 27 yr), EE Cephei (P = 5.6 yr), and OGLELMC-ECL-17782 (P = 13.4 day, identified by Graczyk et al.
2011 in the OGLE LMC EB catalog). The remarkably complicated eclipses of KH 15D are interpreted as a circumbinary
disk eclipsing a pair of stars (Winn et al. 2006). Evidences for
disk-eclipsing nature of young stellar objects YLW 16A and
WL 4 have been presented in Plavchan et al. (2013) and Parks
et al. (2014). An additional disk–eclipse candidate with one occurrence of eclipse was suggested by Mamajek et al. (2012).
A LMC Be star, FTS ID 78.5979.72, which had a sudden transition from no measurable variabilities to eclipsing with short
time variabilities, was reported in Struble et al. (2006).
The period, depth, eclipse duration, and stellar type of OGLELMC-ECL-11893 are similar to those of EE Cephei, suggesting
a similar origin for the peculiar eclipses in both systems. The
H-α line profile is similar to that of ε Aurigae. One important
difference is that the shape of OGLE-LMC-ECL-11893 appears
to be constant while those of EE Cephei, ε Aurigae, OGLELMC-ECL-17782, and KH 15D change between eclipses,
interpreted as precessions of disks. OGLE-LMC-ECL-11893
is exceptional in this respect in that it does not have measurable
precession. Another interesting qualitative difference between
OGLE-LMC-ECL-11893 and EE Cephei is that the former has
relatively short durations of the ingress and egress than the latter.
The disk–eclipse scenario along with other physical properties
of the system will be studied in depth by E. Scott et al. (2014,
in preparation).
Figure 3. Multi-band follow-up light curves of the OGLE-LMC-ECL-11893
BV RI filters (in blue, olive green, red, black) were taken by LCOGT Chilean
1 m telescopes (Dome A: filled circles, Dome B: open circles) during the
2013 eclipse. The OGLE-IV I-band observations (cyan open squares) were
taken in the 2013 and 2011 seasons. The OGLE-III and OGLE-II I-band (black
open pentagons) and V-band (dark green filled pentagons) observations are also
shown for comparison. Observations which were not taken in 2013 are shifted
by periods to match the time of the eclipse in 2013.
(A color version of this figure is available in the online journal.)
other bands. The color variations during the eclipse events may
be related to several physical properties of the system, such
as possible temperature variations across the eclipsed areas,
the opaqueness of the disk as a function of wavelength, the
limb darkening profile of the eclipsed stars, etc. Further highprecision observations are required to accurately measure these
differences and study their implications.
The next two eclipses of OGLE-LMC-ECL-11893 will be
around HJD ∼2,456,818 (2014 June) and ∼2,457,286 (2015
September). However, the former will be practically unobserv3
The Astrophysical Journal, 788:41 (4pp), 2014 June 10
Dong et al.
of Qatar Foundation). C.S. has received funding from the European Union Seventh Framework Programme (FP7/2007-2013)
under grant agreement No. 268421.
The serendipitous identification of OGLE-LMC-ECL-11893
demonstrates the power of the OGLE survey in making discoveries of unanticipated variables, thanks to its long-term, highcadence, and high quality photometry of a large number of stars.
This has implications for planning future large time-domain surveys such as LSST because the overwhelming majority of such
systems in the Galaxy would lie in the inner two quadrants of
the Galactic disk, a region that was excluded in the recently
published LSST planning (see Figure 1 of Gould 2013 and
Figure 4.4 of Abate et al. 2012).
REFERENCES
Abate, A., Aldering, G., Allen, S. W., et al. 2012, arXiv:1211.0310
Bramich, D. M. 2008, MNRAS, 386, L77
Dong, S., Katz, B., & Socrates, A. 2013, ApJL, 763, L2
Dressler, A., Bigelow, B., Hare, T., et al. 2011, PASP, 123, 288
Gałan, C., Mikołajewski, M., Tomov, T., et al. 2012, A&A, 544, A53
Gould, A. 2013, arXiv:1304.3455
Graczyk, D., Soszyński, I., Poleski, R., et al. 2011, AcA, 61, 103
Kloppenborg, B., Stencel, R., Monnier, J. D., et al. 2010, Natur, 464, 870
Koleva, M., Prugniel, P., Bouchard, A., & Wu, Y. 2009, A&A, 501, 1269
Mamajek, E. E., Quillen, A. C., Pecaut, M. J., et al. 2012, AJ, 143, 72
Marshall, J. L., Burles, S., Thompson, I. B., et al. 2008, Proc. SPIE, 7014,
701454
Parks, J. R., Plavchan, P., White, R. J., & Gee, A. H. 2014, ApJS, 211, 3
Plavchan, P., Güth, T., Laohakunakorn, N., & Parks, J. R. 2013, A&A,
554, A110
Silaj, J., Jones, C. E., Tycner, C., Sigut, T. A. A., & Smith, A. D. 2010, ApJS,
187, 228
Slettebak, A. 1979, SSRv, 23, 541
Struble, M. F., Galatola, A., Faccioli, L., Alcock, C., & Cruz, K. 2006, AJ,
131, 2196
Udalski, A., Kubiak, M., & Szymanski, M. 1997, AcA, 47, 319
Udalski, A., Szymanski, M. K., Soszynski, I., & Poleski, R. 2008, AcA,
58, 69
Winn, J. N., Hamilton, C. M., Herbst, W. J., et al. 2006, ApJ, 644, 510
We thank Cullen Blake for stimulating discussion on alternative models and the participants of a lively Institute for Advanced
Study astro-coffee session held in 2011 December. We are
grateful to Alice Quillen and Josh Winn for helpful comments.
S.D. is supported by “the Strategic Priority Research ProgramThe Emergence of Cosmological Structures” of the Chinese
Academy of Sciences (grant No. XDB09000000). The OGLE
project has received funding from the European Research Council under the European Community’s Seventh Framework Programme (FP7/20072013)/ERC grant agreement No. 246678 to
A.U. M.D., R.F.J., K.H., R.A.S., C.S., and Y.T. are supported
by NPRP grant NPRP-09-476-1-78 from the Qatar National
Research Fund (a member of Qatar Foundation). D.M.B.’s contribution to this publication was made possible by NPRP grant #
X-019-1-006 from the Qatar National Research Fund (a member
4